The Spirit of the Sword and Spear

Mark Pearce

From the Norse sagas or the Arthurian cycles, we are used to the concept that the warrior’s weapon has an identity, a name. In this article I shall ask whether some prehistoric weapons also had an identity. Using case studies of La Tène swords, early Iron Age central and southern Italian spearheads and middle and late Bronze Age type Boiu and type Sauerbrunn swords, I shall argue that prehistoric weapons could indeed have an identity and that this has important implications for their biographies, suggesting that they may have been conserved as heirlooms or exchanged as prestige gifts for much longer than is generally assumed, which in turn impacts our understanding of the deposition of weapons in tombs, where they may have had a ‘guardian spirit’ function.

There are many ways in which we can approach the persons and events to which it is connected’. They prehistoric weapons (Pearce 2007): we can study them illustrate this point through Trobriand kula exchange. typologically, to see how their form is related to the In this article I shall take an approach which sequence of types, and we can also use that informa- is related to this latter trend, but rather than try to tion to date them, assigning them to chronological examine the biography of some prehistoric swords horizons. We can examine them functionally, and try and spears, I want to pose the question: was an identity to assess how effective they will have been as weap- attributed to some prehistoric weapons? By using the ons, or perhaps as parade paraphernalia rather than term ‘identity’ I do not mean to argue that prehistoric utilitarian equipment. We can look at use wear and try weapons were regarded as equivalent to humans, but to reconstruct how and for what purpose they were rather that they had some sort of spiritual persona used. Or we can examine them from a metallurgical (which may or may not have been nuanced as regards point of view, looking at how they were made, how aspects such as gender or ethnicity) with its own efficient their edges and points may be, or perhaps specific agency, believed to have its own intention and through chemical analysis trying to reconstruct their volition. This might have been perceived as some sort provenance. More recently other ways of looking at of in-dwelling spirit. material culture have come to the fore, and so for In order to answer the question as to whether example we might examine the ‘biography’ of the an identity was attributed to some weapons we need artefact. to ask how we can know when something has been The biographical approach, which follows the attributed an identity. I would argue that one way is life cycle of an artefact, was proposed by Igor Kopy- where it has been assigned a name, because we give toff, who emphasized that such a biography ‘would names to things to which we attribute a measure of look at … [an object] as a culturally constituted entity, personhood, and therefore agency (Dobres 2000). For endowed with culturally specific meanings, and classi- example, we give names to pets, but not necessarily fied and reclassified into culturally constituted catego- to farm animals. Another way that we attribute an ries’ (Kopytoff 1986, 68); the significance of artefacts identity, or some sort of personhood, to an object is thus changes through time in relation to their context. by giving it eyes, a face or an anthropomorphic form. Gosden and Marshall (1999, 170) add that ‘[n]ot only On the basis of these two observations, I shall do objects change through their existence, but they use some different classes of evidence to argue that often have the capability of accumulating histories, so prehistoric weapons could indeed have identities. First that the present significance of an object derives from I shall use the comparative method, looking at both

Cambridge Archaeological Journal 23:1, 55–67 © 2013 McDonald Institute for Archaeological Research. The online version of this article is published within an Open Access environment subject to the conditions of the Creative Commons Attribution-NonCommercial-ShareAlike licence .55 doi:10.1017/S0959774313000048 Received 22 Jun 2012; Revised 19 Oct 2012; Accepted 16 Nov 2012 Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.234, on 28 Sep 2021 at 09:35:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0959774313000048 Mark Pearce

medieval artefacts and literary sources, myth and epic, 6. general characteristics of the sword (e.g. ‘Long’) and then secondly, I shall work backwards in time, 7. desirable or intended characteristics of the sword looking at particular examples of swords and spears (e.g. ‘Bloodrush’) first from the Iron Age and then from the Bronze Age. 8. the sound of the sword 9. the sharpness of the sword, with reference to Weapon identities particular episodes (e.g. ‘Quern-biter’) 10. words for snakes, wolves and fire (possibly derived The idea that weapons can have names and supernatu- from poetic language in which swords are com- ral powers and consequently agency is a concept that pared to them because of their similar appearance) occurs in medieval epic, such as the twelfth-century 11. abstract concepts (e.g. ‘Agony’, ‘Honour’). Chanson de Roland (171, 2300 to 173, 2344: Roland’s He also notes that some scholars suggest that some sword Durendal could not be broken), and in the swords (such as Flæmingr) are named after Germanic Arthurian cycles. Thomas Malory’s Morte d’Arthur, tribes or ancestral groups, but that other explanations which was written about 1470 (Vinaver 1971, vi), nar- are possible. rates the story of King Arthur’s special sword named We may usefully distinguish two general classes Excalibur, which was given to him by the Lady of the of name in Barnes’s rather nuanced classification: first, Lake (Book 1, 25). This sword was so special that, on names which establish the ownership of the sword his death bed, Arthur instructed one of his knights, Sir (category 1) or which establish the sword’s previous Bedivere, to return it to the waters (Book 21, 5). The biography (categories 2 and 3), and second, names that story was used as an illustration of prehistoric ritual relate to the qualities of the sword itself (categories deposition in water by Richard Bradley (1990, 1–3), 4–11) (cf. Ellis Davidson 1962, 177). but it also attests to the concepts that weapons may Where ownership is denoted, the sword name have names and magical properties and capacities. may not be intended to indicate a specific identity Other weapons in myth cycles or epics could also (as also today we write our names on objects such as have names, and therefore identities (Ellis Davidson books), but Barnes’s classification draws our attention 1962, 82, 102, 151, 177; Barnes 1972; 1982; Kristiansen to the fact that some swords, in addition to having 2002, 329–30): for example, named swords appear names, could also have complex biographies in which, in the Old English poem, Beowulf, in which Beowulf for example, whom they belonged to, or had belonged kills Grendel’s mother with a precious sword called to, or who had made them was important (Barnes’s Hrunting lent him by Unferth (Beowulf 1455–64), and categories 2 and 3; cf. Ellis Davidson 1962, 169–75). also in the Völsunga saga, where Sigurdur Fåvnesbane’s The importance of biographies can be illustrated by sword is called Gramr (‘wrath’: chap. 15) and Fáfnir the two examples from the Norse (i.e. Norwegian and dragon’s sword has the name Hrotti (perhaps translat- Icelandic) sagas: Sigurdur’s Gramr was reforged from able as ‘the audacious one’: chap. 19). Indeed, swords the pieces of his father Sigmund’s sword by the smith, are named in all types of sagas, although infrequently, Regin (Völsunga saga, 15), and in the saga of Gísli and about 100 sword-names are known (Barnes 1972, Súrsson, the sword called Grásíða, which means ‘Grey- col. 544). Barnes notes that not all swords had names sided one’, was broken and then re-forged as a spear and that it seems to have been the custom to give a (Gísla saga Súrssonar 7: Barnes 1972, col. 547). In these sword a name only when it had proved itself in one instances, the name itself does not indicate anything way or another (1972, cols. 544–5). He groups the of their biography but in other cases it could do so, sword names in a number of categories (Barnes 1972, as in the spear Selshefnir (‘the revenger of Sel’) or the cols. 545–6): shield Viljálmsgørð (‘Viljálmr’s handiwork’) (Barnes 1. a personal name in the genitive form with the suffix 1982, col. 283). -nautr ‘gift’ (earlier owner, giver; sometimes this The sword name Curtana, or its variant Curtein, is is ironic, when the personal name is that of the first mentioned in the thirteenth century as denoting original owner who was killed by the subsequent one of the swords carried at English coronations. It owner) was supposed to be the sword of Ogier the Dane, one 2. the nickname of a (previous) owner of Charlemagne’s knights in the Chanson de Roland, 3. the name (or derived from the name) of the smith a weapon with a broken blade which had originally who made it belonged to Tristram (Ditmas 1966). Most of the 4. the general visual appearance of the sword (e.g. coronation regalia were destroyed after the English Ryðfrakki, the red-coated (i.e. rusty) weapon) Civil War during the Commonwealth, and so Curtana, 5. swords with particularly splendid fittings or deco- (also called the ‘Sword of Mercy’) was remade in the ration (e.g. ‘Goldenhilt’) seventeenth century with a broken-off point for the 56

Downloaded from https://www.cambridge.org/core. IP address: 170.106.35.234, on 28 Sep 2021 at 09:35:22, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0959774313000048 The Spirit of the Sword and Spear

coronation of Charles II and this sword is still used in did weapons have identities in prehistory also? The coronations and investitures by the British monarch extent to which the various stories of the Ulster Cycle (Holmes & Sitwell 1972, 12–13). The name thus illus- actually reflect the realities of pre-Christian Ireland trates a long and romantic (but fictitious) biography. is the subject of heated debate, the traditional view As well as swords, Barnes tells us that axes, being that they describe episodes from the time of spears, arrows, shields, helmets and mail-coats all Christ, during the Iron Age of Ireland (Jackson 1964, may have Norse weapon names. For example in the 43–6; Kinsella 1969, ix). Jackson (1964) saw them as Edda, Odin’s spear Gungnir (‘Swayer?’), the helmet a ‘window on the Iron Age’ (as the title of his book Hildigöltr or Hildisvín (‘Battle-boar, -pig’), or Thor’s proclaims) while Aitchison concludes that ‘early Irish hammer Mjöllnir (‘Crusher’) (Barnes 1982, cols. 282–5). epic literature does not constitute a legitimate source These names are found in literature, but some of for the study of pagan Celtic society’ (Aitchison 1987, the events described may have had a historical basis. 113). Specifically looking at the descriptions of swords Indeed, sometimes weapons with names belonged in the Ulster Cycle stories, Mallory (1981, 107) was able to actual historical figures: St Olaf’s axe was named to show ‘there is not a single sword type mentioned after the Norse goddess of the underworld, Hel, and in the Ulster cycle that must be set to the Early Iron Harald hard-ruler’s mail-coat was called Emma, whilst Age’. However he also affirmed that ‘if the Ulster Tales an axe called Kerling (‘Old Woman’) is mentioned in reveal little about the archaeology of Iron Age Ulster, a diploma (Barnes 1982, col. 282). Further confirma- they … do provide us with some idea of what life may tion is given by exceptional archaeological finds. In have been like during the Iron Age and even some Nordic Europe the oldest instance of a weapon name hint of those beliefs that the archaeologist is gener- is raunijar, or ‘Tester’, which is written in runes on ally powerless to recover’ (Mallory & McNeill 1991, the spearhead from Øvre Stabu in Norway, dating to 170), so that it seems reasonable to suggest that it may about ad 150–200 (Barnes 1972, col. 544; Krause 1966, have been the practice to name shields and swords in 75–6, n. 31). There are other early examples of names Iron Age Ireland. Whether or not such a suggestion on spearheads (Krause 1966, 76–82, n. 32 - Dahmsdorf, is reasonable, it is of course very dangerous to use Brandenburg, Germany; n. 33 - Kovel’, Volyns’ka analogies from myth, whether it be Homeric epic, the oblast’, Ukraine; n. 34 - Moos, Gotland, Sweden; and Irish Táin or the Norse sagas, to reconstruct prehistoric n. 35 - Rozwadów, Stalowa Wola, Poland; cf. Barnes reality, so I shall now use archaeological evidence to 1972, col. 544) and on swords from Danish bog-finds ask whether prehistoric weapons also could have an (Ellis Davidson 1962, 42–3, e.g. at Nydam Mose, Søn- identity, indicated by a name or an anthropomorphic derborg, Denmark). form or decoration. Similar examples of named weapons are found in medieval Ireland. The early twelfth-century Irish text La Tène swords Scéla Conchoboir maic Nessa (Tidings of Conchobor mac At least two late La Tène long swords are punch- Nessa) contains a list of shields and swords that were marked with a personal name. A sword from Port(?), kept in King Conchobor’s house — these are identified Canton Bern, Switzerland, datable to early in the by both the name of the weapon and the name of its second half of the first century bc, is stamped ‘Kori- owner (Kinsella 1969, 5). sios’ in Greek script (Fig. 1; Wyss 1956; Livens 1972), This assigning of a name or a biography sug- whilst a sword from Zemplín (Grave 1, tumulus gests the attribution of an identity to weapons. This 8), Slovakia, dated perhaps as late as the end of the is illustrated by the observation that in the early first century bc/beginning of the first century ad, is thirteenth-century, Middle High German epic the marked V]TILICI[O in Latin letters (Pleiner 1993, 80, Nibelungenlied, Siegfried’s sword, Balmung, is almost a 97–8, fig. 11). In neither case is it clear whose name is character in the poem; Hatto (1969, 401) comments that being referred to, and both the smith and the owner ‘in heroic poetry swords are persons’, that is they were have been suggested (Livens 1972; Pleiner 1993, 48 perceived of as having their own identity and agency. n. 5; Wyss 1956), which would fit Barnes’s (1972, col. Thus we may conclude from all these examples that in 545) sword-name categories 1, 2 or 3 (cf. above). It is the Middle Ages of northern Europe weapons could clear however from the foregoing discussion that in have an identity. both cases the name could easily be that of the sword itself, in which case we might hypothesize that an Did prehistoric weapons have identities? identity is being assigned to the sword. It is worth noting that both the Port(?) and the Zemplín swords Having established that weapons had identities in are described as being of good quality (Pleiner 1993, early myth and epic, the question we must ask is: 65, metallographic analysis on pp. 97–8; Wyss 1956, 57

Figure 1. Stamp reading ‘Korisios’ in Greek script on a sword from Port(?) (Canton Bern, Switzerland; source: Wyss 1956, pl. VIIIB).

27) suggesting that well-made swords with fine blades were more likely to be attributed identities. La Tène short swords with an anthropomorphic hilt (Figs. 2 & 3; Clarke & Hawkes 1955; Fitzpatrick 1996) are held to derive from central European Late Hallstatt antennae daggers (some of which have somewhat anthropoid hilts: Clarke & Hawkes 1955, 204, fig. 1:3 & 4). The La Tène anthropomorphic hilted short swords first appear in northern Italy in a Golas- ecca IIIA1 burial at Ca’ Morta, Como (tomb VIII/1926: Negroni Catacchio 1971–72; De Marinis 1981, 56–62, tav. 30:3) dating to the second quarter of the fifth- century bc, and they continue in use over much of ‘Celtic’ Europe until the first centurybc (and in Britain possibly into the first centuryad ), with modifications in the shape of the handle but a remarkably similar design and consistent size. The earliest types are relatively schematic in their representation of the human form, but this becomes more naturalistic through time, Figure 2. La Tène II long sword especially from the second century bc. Moustaches on and anthropomorphic hilted short some of the faces indicate that the figures are male, sword from the North Grimston and Fitzpatrick (1996, 374) argues that they may be (East Yorkshire, England) burial divine, since, apart from heads, full-length human (source: Fitzpatrick 1996, fig. 1). figures are rarely represented in Celtic art (Megaw & Megaw 1989, 21). Fitzpatrick (1996, 376) notes that anthropomorphic hilted short swords are much rarer anthropomorphic hilted short swords have generally than Iron Age long swords, with only around 60–70 been seen to be markers of high status (Filip 1962, known; he argues that they were deposited in less 103), there is not generally a correlation with rich than 1 per cent of all burials. Anthropomorphic hilted burials (Fitzpatrick 1996, 377–8). Fitzpatrick argues short swords are also depicted on Celtic coins (Allen that their short blades and small handles made them 1980, 146, pls. 14:191 & 33:502; Clarke & Hawkes 1955, impracticable as weapons (1996, 376) and on the basis 214–15; De La Tour 1892, pl. XX:6941). of the lunar (and perhaps solar) symbols recognized Hawkes (Clarke & Hawkes 1955, 216) was not on nine examples of mid-late La Tène date (Fig. 3; clear whether the anthropomorphic hilted short Fitzpatrick 1996, 380–85) he posits that they may have swords were ‘felt’ to be human, and wondered had symbolic and ritual functions, particularly associ- whether a divinity might be represented; others ated with counting time, perhaps being used to stab see them as talismanic (Petres 1979, 176). Although victims during human sacrifices (Diodorus Siculus

Figure 3. Anthropomorphic hilted short sword from Figure 4. Long sword with three boar stamps, inlaid Lysice (Moravia, Czech Republic), stamped with a with gold, from Böttstein (Aargau, Switzerland; source: vertical line on the rib and a circle and right facing Fitzpatrick 1996, fig. 6). crescent (source: Fitzpatrick 1996, fig.4).

V, 31, 2 tells us that such victims were stabbed with classes of these stamps: 1) zoomorphic (generally a daggers by the Gauls) (Fitzpatrick 1996, 388–9). boar); 2) mixed (generally a horse and its rider); 3) Between the third and first centuries bc punch- astral; 4) anthropomorphic; and 5) others. Anthropo- marks, sometimes inlaid with gold or copper alloy, morphic punch-marks are the most common, followed appeared on La Tène long swords (Fig. 4), some spears by zoomorphic stamps. Vouga (1923, cols. 36–7) sug- and also a small number of anthropomorphic hilted gested that the stamped symbols might be property short swords. Drack (1954–55, 200–216) identified five marks, or apotropaic, protecting the long sword and

its user, or quality marks for special pieces, whilst with hatched triangles and meanders, circles and dots for Petres (1967–68, 40) they were indicators of high (cat. no. 458); its ‘decoration creates the approxima- prestige as they occur on a relatively low proportion of tion of a human face and headdress’ (Bietti Sestieri & swords. However, as Fitzpatrick (1996, 379–80) points Macnamara 2007, 127). I would suggest that rather out, it is difficult to be sure of their actual prevalence, than a headdress we might prefer to see the crest of a since La Tène sword blades and scabbards are not helmet. Its length (37 cm) and workmanship indicate always well preserved, and corrosion means that it that it is a status symbol as well as a functional weapon is difficult to assess the detail (and thus the degree of (Bietti Sestieri 2006, 514). variability) of the punch-marks. Others have argued Two similar spearheads are present in the collec- that since the stamps were applied by the blacksmith tions of the British Museum, and may also be dated to or armourer that they are maker’s marks, in line with the early phase of the early Iron Age. One has a prov- medieval practice, though some of the weapons on enance of Bari in Puglia, southeast Italy (Fig. 5:2). Bietti which they occur were of average quality, and it is Sestieri and Macnamara (2007, 122, cat. no. 387) who not clear why some blades were stamped and the publish it, suggest that ‘[t]he decoration creates the majority were not (Pleiner 1993, 65; Stead 2006, 49); approximation of a human face’. The circles roughly this would fit with Barnes’s (1972, col. 545) sword represent the eyes and the mouth, whilst the central name category 3. facet suggests a nose. It is relatively long (33 cm). The I would argue that the anthropomorphic hilts of other spearhead (Fig. 5:3), of the same type, came from the short swords and the anthropomorphic stamps the private collection of Count Milano, and is without found on some long swords, spears and short swords provenance, but it may be generically assigned to the signal the identity or personhood attributed to the early Iron Age (Bietti Sestieri & Macnamara 2007, weapons. Whether or not the short swords were 122, cat. no. 388). Here the decoration again shows a actually weapons, their form indicates that they human face, represented by circles, with a headdress symbolized weapons, and their anthopomorphic hilts or helmet-crest indicated by the hatched triangles on seem to indicate the importance of their identity or the blade. It is 29.7 cm long. These two spearheads personhood. It might even be suggested that in some are smaller than the example from Cassino, and their cases a non-anthropomorphic stamp might indicate decoration and workmanship are less refined (Bietti the sword’s name, which would for example suggest Sestieri 2006, 516). that ‘boar’ (Fig. 4) was on occasion used to denote Bietti Sestieri and Macnamara comment on these a sword in mid to late La Tène Europe, in the same three spears (2007, 23) that ‘the human face probably is way as ‘snake’ or ‘wolf’ was in the Nordic sagas (see meant to add a magical or supernatural power to the above, Barnes’s (1972, col. 546) sword-name category spear’. As in the case of the La Tène swords discussed 10). Likewise the lunar or solar symbols may also above, I would add that the face decoration assigns have conferred identity (cf. Barnes’s categories 10 or an identity, or perhaps even spiritual persona with its 11: 1972, col. 546). This identity may have been con- own specific agency, to the spears. ferred at the moment of manufacture, setting stamped A miniature spearhead, also present in the col- swords apart from those which were not chosen to lections of the British Museum, probably comes from bear stamps. a male cremation burial from the Campania region of Italy and is also datable to the early phase of the Early Iron Age spearheads early Iron Age (Bietti Sestieri 2006, 518; Bietti Sestieri & In early Iron Age central and southern Italy, weapons Macnamara 2007, 122, cat. no. 389). It has a decoration appear to be indicators of male status, and while suggesting two eyes and a headdress (Bietti Sestieri swords were rare and therefore seem to indicate high 2006, 518) or helmet-crest. social prestige, spearheads and javelin heads were the Bietti Sestieri (2006, 518–25, figs. 4–9) has shown normal attribute of all or almost all young and adult that the face and headdress iconography continues men (Bietti Sestieri 2006, 507). on spearheads of the typological series in the second A probable high status grave assemblage from phase of the early Iron Age (ninth-eighth century Cassino (Frosinone province), in central Italy, in the bc) in central and southern Italy, with examples British Museum (Fig. 5:1; Bietti Sestieri 2006, 510–14, from Pontecagnano (Salerno), contrada La Rota fig. 2; Bietti Sestieri & Macnamara 2007, 127, cat. nos. (Candidoni, Reggio Calabria), Castellace (Oppido 457–9), datable to the early phase of the early Iron Mamertina, Reggio Calabria), Cairano (Avellino), Age (tenth-early ninth century bc), consists of a type Sala Consilina (Salerno), Marsico Nuovo (Potenza), Cumae sword and two spearheads with a conical Amendolara (Cosenza), Bari, Naples and perhaps socket and foliate blade, one of which is decorated Caracupa (Sermoneta, Latina). Though the decora- 60

Figure 5. Italian Early Iron Age spearheads with face decoration: 1. Cassino (Frosinone; source: Bietti Sestieri & Macnamara 2007, pl. 103, cat. no. 458); 2. Bari (source: Bietti Sestieri & Macnamara 2007, pl. 81, cat. no. 387); 3. From the collection of Count Milano (source: Bietti Sestieri & Macnamara 2007, pl. 81, cat. no. 388).

tion becomes progressively more abstract, she argues Consilina); these may perhaps be part of the decorative that it is likely to have preserved its meaning (Bietti scheme and represent the eyes of the spearhead. Sestieri 2006, 519). Bietti Sestieri (2006, 529) suggests that the richly- Sauerbrunn/Boiu series swords decorated spearheads of this series are rare and likely In a 1966 paper, ‘The origins of the flange-hilted to be markers of status, though she notes that the sword of bronze in Continental Europe’, J.D. Cowen decorative motif is unlikely to have been highly vis- argued that the twin spiral motif on middle Bronze ible. Interestingly, further north, in the contemporary Age type Boiu swords was intended to indicate ‘… male graves of Villanovan central-northern Italy, the a pair of eyes and maybe a brow over them as well. face and headdress iconography does not occur. This These would then be the eyes of the sword itself, or may be because there seems to be a general ritual rather of the spirit within it …’ (Cowen 1966, 294). He prohibition on the deposition of arms in that area; adds that what he calls the ‘magical significance’ of certainly, where spearheads are found (in some high- the design would account for its ‘consistency … over status burials), they are of a different type and do not a wide field for a considerable period’ (Cowen 1966, bear face decoration (Bietti Sestieri 2006, 525). 294). Cowen then goes on to argue that a rapier from Some of the central and south Italian spearheads ‘Hungary’ (which he assigned to his middle Bronze illustrated by Bietti Sestieri have perforations in the Age type Sauerbrunn) carries a depiction of the lower part of their blade (2006, figs. 4:3 - La Rota; 5:1 - whole body of the ‘spirit, or daemon, dwelling in the Castellace; 5:2 - Pontecagnano; 7:2 - Cairano; 7:3 - Sala sword’ (Cowen 1966, 294). Figure 6:1 clearly shows its 61

Figure 6. Cowen’s illustrations of swords with face decoration of his type Sauerbrunn: 1. ‘Hungary’ (source: Cowen 1966, fig. 3:9); 2a. Hochstadt (Germany), face A and 2b. face B (source: Cowen 1966, fig. 3:7 & 8).

outstretched arms, body and legs, and its insect-like illustration of their prevalence and significance. At eyes and serrated teeth set in a large head. I would Olmo di Nogara swords are present in 43 high-status add that the sword from Hochstadt, also illustrated male tombs (out of a total of 149 burials identified by Cowen (1966, fig. 3:7 & 8; assigned to his type osteologically as male), and 19 (Neumann 2009) or 20 Sauerbrunn), shows a similar insect-eyed face, and (De Marinis & Salzani 2005, 393–5, 397–403; Cupitò the faces depicted on either side of the blade can be 2006, 66–7, 71–82, figs. 29, 31–6) of these may be clas- seen in Figure 6:2a and b. sified in the Sauerbrunn/Boiu series. Sixteen of the Swords of the Sauerbrunn/Boiu series have a Sauerbrunn/Boiu swords found at Olmo di Nogara distribution with two focuses, in northeastern Italy show face-type decoration (Fig. 7): in tombs 26, 31, 33, and the Carpathian basin, with outliers in south Ger- 35, 40, 50, 54, 88, 93, 95, 131, 153, 201, 389, 442 and 486 many, Poland, Austria, Slovenia, Croatia and Romania. (Salzani 2005, tav. IV–VIII, X, XI, XIV, XV, XX, XXX, Neumann (2009) lists 99 swords of the series, though XXXVII & XXXIX). It should be noted that as with the some of his attributions are controversial (cf. Cupitò Iron Age spearheads discussed above, in some cases 2006, 66–7, 71–82, not cited by Neumann 2009). As is the decoration tends towards the abstract and can be often the case, there is also disagreement about the recognized as indicating a face only by reference to, detail of their classification and there have been many and comparison with, the more figurative examples. typological schemes since Cowen (e.g. Bianco Peroni 1970, 8–13, 38–51; Kemenczei 1988, 36–42; Schauer Discussion 1971, 20–23, 94–7), but recent studies tend to follow Cowen’s (1966) system (Cupitò 2006, 66–7, 71–82; De We must ask ourselves a number of questions. In the Marinis & Salzani 2005, 393–5, 397–403; Neumann first place, whether there are any alternative hypo 2009). A number of hypotheses have been put forward theses to explain this decoration on Iron Age swords for the origin of the decoration which characterizes and spearheads and middle Bronze Age swords, these swords and these are summarized by Neumann and secondly, whether there is a difference between (2009, 100). weapons with names, and weapons with anthropo- Examining the publications of the 99 Sauer morphic hilts or faces on them, particularly where brunn/Boiu swords listed by Neumann (2009, 112–14), those faces are schematic. Finally, we may ask why I found that at least 46 have decoration that resembles such decoration is not present uniformly, but only on a face. This face decoration is found in both distribu- certain types and on certain examples of those types. tion foci, in northeastern Italy and the Carpathian As regards alternative hypotheses to explain the basin, but also on swords from outlying findspots, decorative motif, it is certainly true that human beings such as in south Germany or Romania. In 21 cases have a tendency to interpret unstructured visual the decoration is missing or partially worn so that it stimuli in meaningful ways (Wertheimer 1923). In paris impossible for me to ascertain the original motif ticular we are prone to ‘see’ faces or eyes for instance and in two further cases (Neumann’s nos. 80 and 81 in wood or in patterns on textiles or wallpaper, when from the River Piave at Colfosco di Sussegana, Treviso, these materials’ stimulus structure have face-like pro northern Italy) there is no illustration published to perties (a phenomenon called pareidolia: Hadjikhani allow the decoration to be assessed. et al. 2009). It might be easy to over-interpret stimuli Forty-six of the swords of the Sauerbrunn/Boiu that may seem to depict faces. But when looking at the series were found in northern Italy and twenty-five of spearheads and swords illustrated in this article, the them have a face-like motif. The large (456 inhumation faces are very striking (see Figs. 5, 6 & 7). It is clearly burials and 61 cremation burials) cemetery of Olmo di impossible to demonstrate conclusively that faces are Nogara (Verona: Salzani 2005) provides an excellent meant, but it does seem evident. 62

Figure 7. Middle Bronze Age swords from the Olmo di Nogara (Verona, Italy) cemetery with face decoration (after Salzani 2005, tav. IV– VIII, X, XI, XIV, XV, XX, XXX, XXXVII, XXXIX).

It could be argued that whereas names are for this decoration are possible, and in some cases individual, faces may be more generic, just indicat- it may again signal the attribution of identity to the ing a human or divine quality, but not necessarily an artefact, but it may also have talismanic or apotropaic individual identity. However, one of the most striking functions. aspects of the depictions of faces on early Iron Age Much of the archaeological evidence that I have spearheads or Bronze Age Sauerbrunn/Boiu series adduced in this article for the ascription of identity to swords is their variation. As can be seen in Figures artefacts — such as stamping and anthropomorphic 5, 6 and 7, the faces are in fact very different. It is my decoration — corresponds to Yvonne Marshall’s (2008) contention indeed that this very variability is a strong concept of ‘inscribed’ objects — they are artefacts argument in favour of their indicating individual that ‘have meaning incorporated into their body in identity. The faces mark them out as different from the course of their making’. In this case ‘[m]eaning is the artefacts of the same type that do not bear faces prescribed in advance of social action. An attempt is (and this distinction is likely to have been conferred made to materially, and thereby socially, fix meaning.’ from the moment of manufacture). (Marshall 2008, 64). On the other hand, unless they That faces occur on both sides of the blades of the were similarly ‘inscribed’, the swords described in early Iron Age spearheads depicted in Figure 5, may the myth cycles or epic outlined above correspond to emphasize their all-seeing nature, so that they literally her category of ‘lived’ objects — that ‘acquire meaning — as Williams puts it when discussing pervasive eye in the context of social action’; this meaning is ‘con- decoration on material from the early seventh-century textual and may change at any time’ (Marshall 2008, ad ship-burial under Mound 1 at Sutton Hoo (Suffolk, 63). Since these swords with names and identities do England) — had eyes in the back of their head (Wil- not necessarily have stamping or anthropomorphic liams 2011, 108). decoration, such ‘lived’ meaning may be difficult to The fact that faces are limited to certain types demonstrate archaeologically. It may, however, be of swords and spearheads, and to certain sub-groups documented by evidence of an extended use-life, for of those types, needs to be explained. It may be that example indicated by repair of an artefact. Thus, the all weapons were considered to have an identity, but scabbard of the Kirkburn (East Yorkshire, England) that only some carried explicit figurative decoration. sword shows signs of several repairs (Stead 2006, Alternatively it may be that particular weapons only 184–5, figs. 85–7; Giles 2008, 61), just as important were considered to have such a quality, and that this swords like Sigurdur’s Gramr or Grásíða were reforged was a result of a specific event in their biography, in in the Norse sagas (see above). this case one would expect the decoration that denoted that identity to have been applied later in the life of Conclusion the artefact (i.e. not at manufacture). In this connec- tion, we may refer again to the Norse sagas, where I have shown that in medieval epic, the Irish Táin Barnes (1972, cols. 544–5) argues that not all swords and the Norse sagas weapons may have names and had names and that it seems to have been the custom supernatural powers and capacities. On occasion to give a sword a name after it had proved itself in such weapons had complex biographies. These char- some way: thus a sword or a spearhead is likely to acteristics are also attested historically and weapon have acquired an identity where it had a significant names written in runes are also known. Early northern biography. A further possibility is that the face and European literature attests that weapons could have eye decoration may be an explicit reference to such an identity. biographical event in the life of a mythical or known It is dangerous to use analogies from myth to weapon of a particular type or provenance: this may reconstruct prehistoric reality and my exploration of explain why in the case of the weapons discussed in the identity of prehistoric weapons has been firmly this paper, the anthropomorphic and face representa- based on archaeological evidence, from both the tions seem to have been applied at manufacture (cf. Bronze and Iron Ages. Two late La Tène long swords Pleiner 1993, 65 for the punch-marks on La Tène weap- are punch-marked with a personal name, which may ons). Finally, it should be noted that not only swords be that of the manufacturer or owner, but may be that and spearheads have anthropomorphic decoration: of the weapon itself. The anthropomorphic hilts of human heads increase ‘in frequency and realism in La Tène short swords — which may have had a sym- the art of the late Iron Age’ (Megaw & Megaw 1989, bolic or ritual rather than combat function — signal 164) and also occur on other artefacts, such as phalerae their identity or personhood as do anthropomorphic (disks), linch-pins and cauldrons (Megaw & Megaw stamps on swords and spears; other punch-marks may 1989, figs. 256–7, 265, 282–6). Various explanations indicate the weapon’s name. Likewise, faces on early 64

Iron Age spearheads from central and southern Italy of pareidolia). Ruth Whitehouse and Jody Joy kindly indicate the identity — perhaps even the personhood commented on previous versions of the paper. Andrew — of the spears. Many years ago, Cowen (1966, 294) Fitzpatrick generously provided his original artwork for suggested that the decoration on swords of the Sauer- Figures 2–4. Permission to reproduce figures was provided by Antiquity Publications, the British Museum Press, the brunn/Boiu series represented ‘the eyes of the sword, Prehistoric Society and Luciano Salzani. Earlier versions of or rather of the spirit within it …’. My wide-ranging this paper were given at the session on ‘New approaches on discussion suggests that Cowen’s intuition was both studying weaponry of the European Bronze Age’ (organ- plausible and credible, and that in prehistoric Europe, ized by Marion Uckelmann and Marianne Mödlinger) at as in later periods, both swords and spearheads could the 15th Annual Meeting of the European Association of have some sort of identity, and it may have been Archaeologists in Riva del Garda, Italy, 15–20 September believed that they were indwelt by spirits or daemons. 2009, at a seminar on prehistoric Italy held at the University Does this have any consequences? If we accept of Nottingham on 17 February 2010, and at a seminar at the that there was a belief that some Bronze Age Sau- University of Padua, Italy, 17 June 2010; I am very grateful to all those who commented in the discussion on those erbrunn/Boiu series swords, Italian early Iron Age occasions. This paper was revised after refereeing at the spearheads and La Tène swords and spears had an Römisch-Germanische Kommission in Frankfurt: I am very identity, or indeed an in-dwelling spirit, like those of grateful to the Director, Svend Hansen, for hospitality and the medieval myths and epic, an identity that might bibliographic advice. None of the above have any respon- relate to their biographies, then they might have been sibility for any errors or omissions in the paper. conserved as heirlooms or exchanged as prestige gifts for much longer than is generally assumed. Such Mark Pearce practices are difficult to document archaeologically, Department of Archaeology because we generally date artefacts on the basis of University of Nottingham their associations, but they have been argued for University Park weapons circulating in Anglo-Saxon society (Härke Nottingham 2000) and also for the sword in tomb 42 at Olmo NG7 2RD di Nogara and the type Sauerbrunn sword from UK Sauerbrunn, Burgenland, Austria, by analogy with Email: [email protected] Odysseus’ boars’ tusk helmet (Iliad 10, 260–71) which was passed across the generations by heroic gift- References exchange and inheritance (Cupitò 2005, 242). Likewise the sword Ættartangi passed down the generations and Aitchison, N.B., 1987. The Ulster Cycle: heroic image and appears in both Vatnsdœla Saga and Grettis Saga (Ellis historical reality. Journal of Medieval History 13, 87–116. Davidson 1962, 171–2) and Barnes comments that the Allen, D.F., 1980. The Coins of the Ancient Celts, ed. D. Nash. Norse name for this sword is a name which denotes Edinburgh: Edinburgh University Press. Barnes, M., 1972. Sverdnavn, in Kulturhistorisk leksikon for that the sword was meant to stay in a family for many nordisk middelalder fra vikingetid til reformationstid, vol. generations (1972, col. 545). These examples suggest 17. Copenhagen: Rosenkilde og Bagger, cols. 544–7. that some weapons may have circulated a long time Barnes, M., 1982. Våpennavn, in Kulturhistorisk leksikon for after their manufacture (cf. Whitley 2002), which in nordisk middelalder fra vikingetid til reformationstid, vol. turn may have serious implications for the use of such 20. 2nd edition. Copenhagen: Rosenkilde og Bagger, weapons in typological dating schemes and for our cols. 282–5. understanding of their deposition in tombs. Artefacts Bianco Peroni, V., 1970. Le spade nell’Italia Continentale. (Prä- which because of their special identity had acquired a historische Bronzefunde IV, 1.) Munich: Beck. complex biography and to which stories had become Bietti Sestieri, A.M., 2006. Fattori di collegamento- inter regionale nella prima Età del Ferro: indizi di attached, concerning for example their previous un’ideologia condivisa, legata alle armi, dal Lazio owners or their prowess in combat, are likely to have meridionale alla Puglia. Rivista di Scienze Preistoriche had specific roles in social relations, and this special, 56, 505–33. powerful, identity may have conferred on them a Bietti Sestieri, A.M. & E. Macnamara, 2007.Prehistoric Metal ‘guardian spirit’ function when they were deposited. Artefacts from Italy (3500–720bc) in the British Museum. (Research Publication 159.) London: British Museum. Acknowledgements Bradley, R., 1990. The Passage of Arms: an Archaeological Analy- sis of Prehistoric Hoards and Votive Deposits. Cambridge: I am greatly indebted for their invaluable help with the Nor- Cambridge University Press. dic examples to Judith Jesch and Torill Christine Lindstrøm Clarke, R.R. & C.F.C. Hawkes, 1955. An iron anthropoid (who also advised me on the psychological phenomenon sword from Shouldham, Norfolk with related

continental and British weapons. Proceedings of the Jackson, K.H., 1964. The Oldest Irish Tradition: a Window on Prehistoric Society 21, 198–227. the Iron Age. (The Rede Lecture 1964.) Cambridge: Cowen, J.D., 1966. The origins of the flange-hilted sword Cambridge University Press. of bronze in continental Europe. Proceedings of the Kemenczei, T., 1988. Die Schwerter in Ungarn I (Griffplatten-, Prehistoric Society 32, 262–312. Griffangel- und Griffzungenschwerter). (Prähistorische Cupitò, M., 2005. La necropoli dell’età del bronzo a Gam- Bronzefunde IV, 6.) Munich: Beck. baloni di Povegliano. Rilettura e reinterpretazione Kinsella, T., 1969. The Táin: Translated from the Irish Epic Táin dei dati ottocenteschi alla luce dei nuovi documenti Bó Cuailnge. Dublin: Dolmen Press. [Reprinted Oxford: d’archivio, in Archeologia e idrografia del Veronese a Oxford University Press, 1970.] cent’anni dalla deviazione del fiume Guà (1904–2004). Il Kopytoff, I., 1986. The cultural biography of things: com- Museo Archeologico di Cologna Veneta e le prime ricerche moditization as process, in The Social Life of Things: archeologiche nella pianura veronese, eds. G. Leonardi Commodities in Cultural Process, ed. A. Appadurai. & S. Rossi. (Saltuarie dal laboratorio del Piovego 6.) Cambridge: Cambridge University Press, 64–91. Cologna Veneta: Ambrosini, 197–265. Krause, W., 1966. Die Runeninschriften im älteren Futhark. Cupitò, M., 2006. Tipocronologia del Bronzo medio e recente (Abhandlungen der Akademie der Wissenschaften tra l’Adige e il Mincio sulla base delle evidenze funerarie. in Göttingen, Philologisch-Historische Klasse; Dritte (Saltuarie dal laboratorio del Piovego 7.) Padua: Folge, 65.) Göttingen: Vandenhoeck und Ruprecht. Imprimitur. Kristiansen, K., 2002 The tale of the sword — swords and De Marinis, R., 1981. Il periodo Golasecca III A in Lombardia. swordfighters in Bronze Age Europe. Oxford Journal Studi Archeologici 1, 41–303. of Archaeology 21(4), 319–32. De Marinis, R.C. & L. Salzani, 2005. Tipologia e cronologia De La Tour, H. , 1892. Atlas de monnaies gauloises. Paris: Plon, dei materiali, in La necropoli dell’età del Bronzo all’Olmo Nourrit et Cie. [Reprinted Lund: Humphries, 1965.] di Nogara, ed. L. Salzani. (Memorie II serie, Sezione Livens, R.G., 1972. Who was Korisios? Antiquity 46(181), Scienze dell’Uomo 8.) Verona: Museo Civico di Storia 56–7. Naturale di Verona, 391–448. Mallory, J.P., 1981. The sword of the Ulster Cycle, in Studies Ditmas, E.M.R., 1966. The Curtana or Sword of Mercy. in Early Ireland: Essays in Honour of M.V. Duignan, ed. Journal of the British Archaeological Association 29 (third B.G. Scott. Belfast: Association of Young Irish Archaeo series), 122–33. logists, 99–114. Dobres, M.-A., 2000. Technology and Social Agency. Oxford: Mallory, J.P. & T.E. McNeill, 1991. The Archaeology of Ulster Blackwell. from Colonization to Plantation. Belfast: The Institute Drack, W., 1954–55. Ein Mittellatèneschwert mit drei Gold- of Irish Studies, The Queen’s University of Belfast. marken von Böttstein (Aargau).Zeitschrift für Schwei- Marshall, Y., 2008. The social lives of lived and inscribed zerische Archäologie und Kunstgeschichte 15, 193–235. objects: a Lapita perspective. Journal of the Polynesian Ellis Davidson, H.R., 1962. The Sword in Anglo-Saxon England: Society 117(1), 59–101. Its Archaeology and Literature. Oxford: Clarendon Press. Megaw, M.R. & J.V.S. Megaw, 1989. Celtic Art: from its Begin- Filip, J., 1962. Celtic Civilization and its Heritage. Prague: New nings to the Book of Kells. London: Thames and Hudson. Horizons. Publishing House of the Czechoslovak Negroni Catacchio, N., 1971–72. Spade con impugnatura Academy of Sciences. pseudo-antropoide nell’area della cultura di Gola- Fitzpatrick, A.P., 1996. Night and day: the symbolism of secca. Rapporti tra il La Tène e il Golasecca III A. astral signs on later Iron Age anthropomorphic short Sibrium 11, 113–31. swords. Proceedings of the Prehistoric Society 62, 373–98. Neumann, D., 2009. Bemerkungen zu den Schwertern der Giles, M., 2008. Seeing red: the aesthetics of martial objects Typenfamilie Sauerbrunn-Boiu-Keszthely, in Alpen, in the British and Irish Iron Age, in Rethinking Celtic Kult und Eisenzeit: Festschrift für Amei Lang zum 65. Art, eds. D. Garrow, C. Gosden & J.D. Hill. Oxford: Geburtstag, eds. J.M. Bagley, C. Eggl, D. Neumann Oxbow Books, 59–77. & M. Schefzik. (Internationale Archäologie: Studia Gosden, C. & Y. Marshall, 1999. The cultural biography of honoraria 30.) Rahden/Westf.: Leidorf, 97–114. objects. World Archaeology 31(2), 169–78. Pearce, M., 2007. Bright Blades and Red Metal: Essays on North Hadjikhani, N., K. Kveraga, P. Naik & S.P. Ahlfors, 2009. Italian Prehistoric Metalwork. (Specialist Studies on Italy Early (M170) activation of face-specific cortex by face- 14.) London: Accordia Research Institute. like objects. NeuroReport 20(4), 403–7. Petres, É.F., 1967–68. Zum Problem der gestempelten La Härke, H., 2000. The circulation of weapons in Anglo-Saxon Tène-Zeitlichen Schwerter. Alba Regia. Annales Musei society, in Rituals of Power: from Late Antiquity to the Stephani Regis 8–9, 35–42. Early Middle Ages, eds. F. Theuws & J.L. Nelson. Petres, É.F., 1979. Some remarks on anthropoid and Leiden: Brill, 377–99. pseudoanthropoid hilted daggers in Hungary, in Hatto, A.T., 1969. The Nibelungenlied, trans. A.T. Hatto. Les mouvements celtiques du Ve au Ier siècle avant notre Harmondsworth: Penguin. ère: Actes du XXVIIIe colloque organisé à l’occasion du Holmes, M. & H.D.W. Sitwell, 1972. The English Regalia: Their IXe Congrès International des Sciences Préhistoriques et History, Custody and Display. London: Her Majesty’s Protohistoriques, Nice, le 19 septembre 1976, eds. P.-M. Stationery Office. Duval & V. Kruta. Paris: Editions du Centre National

de la Recherche Scientifique, 171–8. and fallacies in the study of Late Bronze Age and Early Pleiner, R., 1993. The Celtic Sword, with contributions by B.G. Iron Age warrior graves. Cambridge Archaeological Scott. Oxford: Clarendon Press. Journal 12(2), 217–32. Salzani, L., 2005. La necropoli, in La necropoli dell’età del Williams, H., 2011. The sense of being seen: ocular effects at Bronzo all’Olmo di Nogara, ed. L. Salzani. (Memorie II Sutton Hoo.Journal of Social Archaeology 22(1), 99–121. serie, Sezione Scienze dell’Uomo 8.) Verona: Museo Wyss, R., 1956. The sword of Korisios. Antiquity 30(117), Civico di Storia Naturale di Verona, 9–388. 27–8. Schauer, P., 1971.Die Schwerter in Süddeutschland, Österreich und der Schweiz I (Griffplatten-, Griffangel- und Griff- Author biography zungenschwerter). (Prähistorische Bronzefunde IV, 2.) Munich: Beck. Mark Pearce is Associate Professor in Archaeology at the Stead, I.M., 2006. British Iron Age Swords and Scabbards. University of Nottingham. He specializes in north Italian London: British Museum Press. prehistory, particularly early copper metallurgy and mining Vinaver, E. (ed.), 1971. Malory: Works. 2nd edition. Oxford: and the early Neolithic. He is author of Bright Blades and Red Oxford University Press. Metal: Essays on North Italian Prehistoric Metalwork, Il territorio Vouga, P., 1923. La Tène: Monographie de la station publiée au di Milano e Pavia tra Mesolitico e Prima età del Ferro: Dalla nom de la Commission des Fouilles de la Tène. Leipzig: carta archeologica alla ricostruzione del paesaggio and Materiali Karl W. Hiersemann. preistorici: Cataloghi dei Civici Musei di Pavia, I. He recently Wertheimer, M., 1923. Untersuchungen zur Lehre von der co-edited The Death of Archaeological Theory? and Ligurian Gestalt. Psychologische Forschung4, 301–50. Landscapes: Studies in Archaeology, Geography and History. Whitley, J., 2002. Objects with attitude: biographical facts

Editor Neil Hopkinson, Trinity College, Cambridge, UK Roger Rees, University of St Andrews, UK

The Classical Review publishes informative reviews from leading The Classical Review scholars on new work covering the literatures and civilizations is available online at: http://journals.cambridge.org/car of ancient Greece and Rome. Publishing over 300 high quality reviews and 100 brief notes every year, The Classical Review is an indispensable reference tool, essential for keeping abreast with current classical scholarship.

To subscribe contact Customer Services

in Cambridge: Phone +44 (0)1223 326070 Fax +44 (0)1223 325150 Email [email protected]

in New York: Phone +1 (845) 353 7500 Fax +1 (845) 353 4141 Email [email protected] Price information is available at: http://journals.cambridge.org/car

Free email alerts Keep up-to-date with new material – sign up at http://journals.cambridge.org/alerts

For free online content visit: http://journals.cambridge.org/car